MOTIVATION

The capacity of the ocean to absorb anthropogenic CO₂ is controlled largely by its ability to buffer the pH and carbonate ion concentration [CO₃^2-] through several mechanisms or feedbacks, each of which operates on different time scales (Ridgwell and Zeebe, 2005). On time scales of thousands of years, the most important feedback involves the dissolution of carbonate sediments on the seafloor, which can restore pH and [CO₃^2-], and thus allow the ocean to absorb additional carbon (Archer et al., 1997). The majority of carbonate sediment, however, resides on the deep sea floor. As a consequence, the process of buffering through carbonate dissolution is paced partly by the rate at which ocean overturn can deliver acidified water to the deep sea. Because the rate of overturn, roughly 10³ y, is comparatively slow relative to the rate of emissions, CO₂ is rapidly accumulating in the atmosphere. In theory, with several mixing cycles, the bulk of the anthropogenic carbon will slowly propagate into the deep sea, and trigger widespread dissolution of carbonate, thereby sequestering CO₂ at a predictable rate. As the system drifts further from present conditions, however, the rate of sequestration could deviate from predictions. For one, the rate of oceanic overturn should slow as high-latitude surface waters become warmer and fresher, which would reduce absorption. Also, the chemical erosion of seafloor carbonate is limited by the rate of resupply of fresh carbonate sediment via bioturbation (Archer and Maier-Reimer, 1994). Should the organisms responsible for bioturbation disappear, say for a lack of oxygen or warmer temperatures, then the resupply of carbonate for weathering should cease as well. On even shorter time scales, the export production of carbon, which is controlled by factors such as production and flux of calcareous algae, could change with acidification and climate change. Each of these potential surprises increases the degree of uncertainty in forecasts of ocean carbon absorption and sequestration.

To evaluate our theoretical understanding of the complex processes that govern the carbon cycle, and to test the sensitivity of coupled climate/biogeochemical models to extreme forcing, researchers are increasingly turning to Earth’s past to study periods of rapid greenhouse warming. Of particular interest is a transient global warming event characterized by a carbon cycle perturbation at the Paleocene-Eocene boundary 55 Mya referred to as the Paleocene-Eocene Thermal Maximum (PETM). As evidenced by a prominent carbon isotope excursions (CIE) and global deep-sea dissolution horizon (Fig. 1), the PETM was accompanied by the release of ~4000 PgC over a period significantly shorter than the residence time of carbon in the ocean (<10 ky; Zachos et al., 2005). Interestingly, this carbon release, which left an indelible imprint in the sedimentary and fossil record, may not have been as large as the potential input of carbon from fossil-fuel burning.

The last five years have seen a surge in our understanding of the dynamics of the PETM event including far more precise constraints of the range of climatic, biotic and biogeochemical responses on both short and long time-scales. Significant discoveries have been made by an international team funded through NSF-Biocomplexity who have taken advantage of the recovery of spectacular new marine and terrestrial records of the event (e.g., Zachos et al., 2003; 2005; Bowen et al., 2004, in press; Wing et al., 2005; see http://es.ucsc.edu/~silab/biocomplex/index.htm for more information) as well as the development of powerful new geochemical proxies (Zachos et al., submitted). In addition, we now recognize that the PETM was not unique: a second event, referred to as ELMO (Lourens et al., 2005), occurred some two million years later and exhibits many of the same features of the PETM. While the pace of discovery has been rapid, the exact cause(s) of these events is still a matter of debate and we have been unable to fully quantify the processes responsible for evaluation of the long-term carbon cycle recovery rates and modes of sequestration, and to test models of how biogeochemical feedbacks work to buffer the ocean and sequester carbon. In particular, uncertainties in chronology prevent us from fully interpreting the rates of carbon addition during the onset of the events, and especially, removal during the recovery. Thus, our understanding of the dynamics of these events and potential implications for present and future carbon cycle perturbation is less than complete.

Here we propose a 4-year investigation focused on the dynamics of ocean carbon uptake and sequestration during two early Eocene hyperthermals, with a particular emphasis on the PETM. This investigation brings together an international team of experts in ocean and carbon cycle modeling, marine and sedimentary geochemistry, and paleoceanography, to jointly address, through data/model comparison, several key questions regarding the nature of the carbon cycle perturbations during these events; 1) What were the mass, rate, and origin of carbon released during the events? 2) What were the rates of sequestration and recovery and what biogeochemical feedbacks came into play? and 3) How did associated extreme changes in ocean carbonate chemistry affect planktonic calcifiers?